Remote sensing

Appreciate the gravity

Article metrics

Two satellites will soon be launched that can measure annual variations in the Earth's gravity due to mass changes equivalent to 1 cm of water over 250,000 km2 — an area smaller than the Caspian Sea. This is gravity measurement of unprecedented accuracy. It will affect nearly all areas of study of the Earth, with the greatest advances expected in the study of ocean dynamics, continental water-table variations, sea-level rise, glaciology, and postglacial rebound. These possible applications were discussed at a meeting last monthFootnote 1 and have been addressed in a National Research Council (NRC) report1.

The first mission, CHAMP, is being developed in Germany with cooperation from the United States and France, and should be launched in 1999. CHAMP is a low-Earth orbiter whose main purpose is to study the Earth's magnetic field. However, it will be equipped with three global-positioning-system (GPS) receivers looking fore, aft and up. These receivers will be used to measure atmospheric refractivity (as GPS satellites go behind the Earth) and to refine our picture of the large-scale gravity field.

But the second mission, the Gravity Recovery and Climate Experiment (GRACE), to be launched by the US space agency NASA in 2001, is expected to provide the more detailed view of the changes in the Earth's gravity field. GRACE will be a pair of satellites, separated by a few hundred kilometres, and orbiting at an altitude of about 600 km for 3-5 years. These satellites will accurately measure changes in their separation, produced as they orbit the Earth following the bumps in its gravity field (Fig. 1). If the predicted measurement accuracy is realized, these satellites will give us a remarkably precise view of the Earth's gravity field and its fluctuation.

Figure 1: Sinuous grace: a simulation of the changes in separation between the two GRACE satellites, for two different mass changes on the Earth's surface.

The satellites are assumed to be 400 km up in near-polar orbits, and at day zero, 1 cm of water is added over 250,000 km2 at the Equator (red curve) and at 30° latitude (green). The deflections produced are much more than the expected measurement accuracy of 0.001 mm; the real problem will be subtracting the signal produced by the steady part of Earth's gravity field, a separation change of about 1 km with a periodic pattern like that in the insert.

The gravity field provides a record of the Earth's mass distribution, and so can be used to understand the structure and dynamics needed to maintain that distribution. For the more fluid portions of the Earth, gravity measurements can be used to sense motions of mass. Calculating the mass distribution and dynamics from the gravity field is not a straightforward problem, but, through a combination of spatial and temporal analyses, insight can be gained into the processes that control these dynamics. For example, the distribution of mass in the mantle can be used to measure the vigour of mantle convection.The reliability of these inferences depends on the accuracy, spatial resolution, temporal resolution and duration of the gravity measurements.

The size of the mass changes GRACE can see depends on a number of factors. Perhaps counter-intuitively, it will be most sensitive to spread-out changes: the larger the area over which the mass change occurs, the larger the perturbations to the satellites' orbits. Also, the longer a mass change exists the more accurately it can be measured, because of the increased averaging time.

The NRC report1 details the sensitivity of an imagined mission similar to GRACE by expressing mass changes as equivalent thicknesses of water over regions of different sizes, and over timescales of 90 days and up. For example, the groundwater level of the High Plains aquifer in the Great Plains region of the United States has fallen more than 30 m in places over the past 40 years2. This aquifer covers 750,000 km2. Gradual changes over regions a third of this size should be measurable to an accuracy of < 5 mm yr−1 over a five-year mission; and annual variations of 10-30 mm over these same regions should also be detectable. Such integrated estimates of changes in the water volume would be hard to obtain with conventional hydrological measurements.

Over the world's ocean, sea-level changes of < 1 mm yr−1 should be measurable, and, because the gravity mission will also detect the mass changes of the oceans, it will be possible to separate the thermal-expansion term of sea-level change from the water-mass term. Also, it should be possible to determine where the additional water is coming from — the current state of change of the world's large ice sheets, for example, is still poorly understood.

Interpreting changes in gravity is not without its problems. The mass distribution that generates a given external potential field is not unique. But most of the mass variations on timescales less than years are expected to arise from changes near the Earth's surface, and these can be inferred uniquely as surface mass densities. The atmosphere must be taken into account in the interpretation of these surface density changes, because it redistributes large masses and it is the most dynamic of the fluids in and on the Earth. In many regions of the world, existing models for atmospheric mass variations will probably be adequate to remove the atmosphere's contributions — but that will not be the case everywhere.

In addition to gravity changes, the next generation of gravity missions will greatly improve our knowledge of the static portion of the field. When combined with ancillary data (seismology, geology and laboratory measurements of materials at high pressure and temperature), the static gravity field should improve our understanding of the structure and evolution of the crust and lithosphere, and the processes involved in mantle dynamics and the deep structure of the Earth.


  1. 1.

    * American Geophysical Union Fall Meeting, San Francisco, 8-12 December 1997.


  1. 1

    Satellite Gravity and the Geosphere (Natl Acad. Press, Washington DC, 1997).

  2. 2

    Dugan, J. T. & Cox, D. A. US Geological Survey, Water Resources Investigations Report 94-4157 (1994).

Download references

Author information

Rights and permissions

Reprints and Permissions

About this article

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.